DESCRIPTION: In this application our overall hypothesis is that HIV-1 vectors can be developed that are both safe and effective for delivery of genes to protect cells from HIV-1 infection. The use of gene therapy in humans requires, first, a better understanding of the parameters necessary for efficient transduction, transplant, marking and gene expression. Second, we must utilize this information to model gene therapy for specific human diseases in this case, H IV-1 disease. The primary advantage of HIV-1 based vectors is that they are derived from a virus which has evolved to efficiently infect human cells. However, this property is also the basis of the major reservation regarding the safety of these vectors. Since we believe there are many advantages to the use of lentiviral vectors, particularly in the context of HIV-1 disease, it is critical that if lentiviral vectors are to move forward into the human clinical arena, that their substantial theoretical advantages be documented in primate model systems. We propose to utilize the SCID-hu model for human CD34+ T-progenitor cell transplant and the rhesus macaque CD34+ cell autologous transplant system. As outlined in our Preliminary Studies, we successfully developed protocols to exploit these model systems to better understand the properties of HIV-1 vectors. We also developed a novel inducible HIV-1 vector (DAt1 ru) which itself inhibits HIV-1 replication in the absence of other anti-HIV-1 genes. We propose to understand the mechanism of action of this vector and then test this vector using the two primate model systems described above. The specific aims are: 1) Determine the mechanism by which the inducible HIV-1 based vector inhibits HIV-1 replication. 2) Develop further understanding of the requirements for efficient HIV-1 vector transduction, transplantation, multi-lineage marking, gene expression, and immune response. 3) Based upon the results of Aims 1 and 2, test the effectiveness of the best anti-HIV gene therapeutic vectors to inhibit HIV-1 1/SIV replication in in vivo model systems of SCID-hu mice and rhesus macaque.